Scrollspy not working on docs secondary nav when running locally

[ackernaut]

Steps to reproduce this issue:

1. Pull latest from master branch
2. Run jekyll serve
3. Go to http://localhost:9001/

As you scroll the sections the top list item will have the active class but the secondary list items are not targeted.

<li class="active">
  <a href="#dropdowns">Dropdowns</a>
  <ul class="nav">
    <li><a href="#dropdowns-example">Example</a></li>
    <li><a href="#dropdowns-alignment">Alignment options</a></li>
    <li><a href="#dropdowns-headers">Headers</a></li>
    <li><a href="#dropdowns-disabled">Disabled menu items</a></li>
  </ul>
</li>

[XhmikosR]

Confirmed.

78cac11 seems to have caused it from #13541.

[XhmikosR]

/CC @fat

[cvrebert]

I guess Affix needs some more unit tests.

[XhmikosR]

Yeah. I noticed it too yesterday, but I thought my local tree was messed up so I didn't dig into this more.

[XhmikosR]

What's interesting is that I don't see the issue present in my xhmikosr-js-style branch from #13388.

[cvrebert]

Another reason to merge in the style improvements then! 😄

[XhmikosR]

OK, I was tricked. Basically if you open http://127.0.0.1:9001/components/#x then everything is fine for x but not for the rest.

[XhmikosR]

So it's not the above patch I mentioned that's causing the problem. I'll try to pinpoint the faulty commit.

[ackernaut]

@XhmikosR Yeah, I see what you mean. The parent nav section will work if I load the page with a sub-nav item:

http://localhost:9001/components/#dropdowns-headers

[XhmikosR]

Yeah and it makes things harder to pinpoint the cause :(

I have reverted all relevant changes after 3.1.1 and I still can't make it to work. The 3.1.1 tag works fine, so it must be something really weird...

[mdo]

The parent links get .active for me in Safari, but I don't see the subnav links getting the class. I checked the scrollspy plugin and nothing looks off there—not a lot of commits against that file.

[XhmikosR]

OK, I have found the commit that caused it. It's 0f366e0 from #12337.

[fat]

hm… im not seeing this problem in chrome…

[cvrebert]

I can't repro with 0f366e0 even.
@XhmikosR Can you try and repro this with the latest code?

[mdo]

Yeah, working for me now in latest master.

[XhmikosR]

Confirmed.